* Copyright (c) 2003, 2010, Oracle and/or its affiliates. All rights reserved.
* Use is subject to license terms.
*
- * Copyright (c) 2011, 2014, Intel Corporation.
+ * Copyright (c) 2011, 2015, Intel Corporation.
*/
/*
* This file is part of Lustre, http://www.lustre.org/
#include <linux/module.h>
#include <linux/statfs.h>
+#include <linux/time.h>
#include <linux/types.h>
#include <linux/version.h>
#include <linux/mm.h>
struct kmem_cache *ll_file_data_slab;
-static struct list_head ll_super_blocks = LIST_HEAD_INIT(ll_super_blocks);
-static DEFINE_SPINLOCK(ll_sb_lock);
-
#ifndef log2
#define log2(n) ffz(~(n))
#endif
class_uuid_unparse(uuid, &sbi->ll_sb_uuid);
CDEBUG(D_CONFIG, "generated uuid: %s\n", sbi->ll_sb_uuid.uuid);
- spin_lock(&ll_sb_lock);
- list_add_tail(&sbi->ll_list, &ll_super_blocks);
- spin_unlock(&ll_sb_lock);
-
sbi->ll_flags |= LL_SBI_VERBOSE;
#ifdef ENABLE_CHECKSUM
sbi->ll_flags |= LL_SBI_CHECKSUM;
ENTRY;
if (sbi != NULL) {
- spin_lock(&ll_sb_lock);
- list_del(&sbi->ll_list);
- spin_unlock(&ll_sb_lock);
if (!list_empty(&sbi->ll_squash.rsi_nosquash_nids))
cfs_free_nidlist(&sbi->ll_squash.rsi_nosquash_nids);
if (sbi->ll_cache != NULL) {
OBD_CONNECT_FLOCK_DEAD |
OBD_CONNECT_DISP_STRIPE | OBD_CONNECT_LFSCK |
OBD_CONNECT_OPEN_BY_FID |
- OBD_CONNECT_DIR_STRIPE;
+ OBD_CONNECT_DIR_STRIPE |
+ OBD_CONNECT_BULK_MBITS |
+ OBD_CONNECT_SUBTREE;
#ifdef HAVE_LRU_RESIZE_SUPPORT
if (sbi->ll_flags & LL_SBI_LRU_RESIZE)
if (sbi->ll_flags & LL_SBI_RMT_CLIENT)
data->ocd_connect_flags |= OBD_CONNECT_RMT_CLIENT_FORCE;
+ /* always ping even if server suppress_pings */
+ if (sbi->ll_flags & LL_SBI_ALWAYS_PING)
+ data->ocd_connect_flags &= ~OBD_CONNECT_PINGLESS;
+
data->ocd_brw_size = MD_MAX_BRW_SIZE;
err = obd_connect(NULL, &sbi->ll_md_exp, obd, &sbi->ll_sb_uuid, data, NULL);
GOTO(out_md_fid, err = -ENODEV);
}
+ /* pass client page size via ocd_grant_blkbits, the server should report
+ * back its backend blocksize for grant calculation purpose */
+ data->ocd_grant_blkbits = PAGE_SHIFT;
+
data->ocd_connect_flags = OBD_CONNECT_GRANT | OBD_CONNECT_VERSION |
OBD_CONNECT_REQPORTAL | OBD_CONNECT_BRW_SIZE |
OBD_CONNECT_CANCELSET | OBD_CONNECT_FID |
OBD_CONNECT_EINPROGRESS |
OBD_CONNECT_JOBSTATS | OBD_CONNECT_LVB_TYPE |
OBD_CONNECT_LAYOUTLOCK |
- OBD_CONNECT_PINGLESS | OBD_CONNECT_LFSCK;
+ OBD_CONNECT_PINGLESS | OBD_CONNECT_LFSCK |
+ OBD_CONNECT_BULK_MBITS;
+
+ if (!OBD_FAIL_CHECK(OBD_FAIL_OSC_CONNECT_GRANT_PARAM))
+ data->ocd_connect_flags |= OBD_CONNECT_GRANT_PARAM;
if (!OBD_FAIL_CHECK(OBD_FAIL_OSC_CONNECT_CKSUM)) {
/* OBD_CONNECT_CKSUM should always be set, even if checksums are
if (sbi->ll_flags & LL_SBI_RMT_CLIENT)
data->ocd_connect_flags |= OBD_CONNECT_RMT_CLIENT_FORCE;
+ /* always ping even if server suppress_pings */
+ if (sbi->ll_flags & LL_SBI_ALWAYS_PING)
+ data->ocd_connect_flags &= ~OBD_CONNECT_PINGLESS;
+
CDEBUG(D_RPCTRACE, "ocd_connect_flags: "LPX64" ocd_version: %d "
"ocd_grant: %d\n", data->ocd_connect_flags,
data->ocd_version, data->ocd_grant);
mutex_unlock(&sbi->ll_lco.lco_lock);
fid_zero(&sbi->ll_root_fid);
- err = md_getstatus(sbi->ll_md_exp, &sbi->ll_root_fid);
+ err = md_get_root(sbi->ll_md_exp, get_mount_fileset(sb),
+ &sbi->ll_root_fid);
if (err) {
CERROR("cannot mds_connect: rc = %d\n", err);
GOTO(out_lock_cn_cb, err);
*flags &= ~tmp;
goto next;
}
+ tmp = ll_set_opt("context", s1, 1);
+ if (tmp)
+ goto next;
+ tmp = ll_set_opt("fscontext", s1, 1);
+ if (tmp)
+ goto next;
+ tmp = ll_set_opt("defcontext", s1, 1);
+ if (tmp)
+ goto next;
+ tmp = ll_set_opt("rootcontext", s1, 1);
+ if (tmp)
+ goto next;
tmp = ll_set_opt("remote_client", s1, LL_SBI_RMT_CLIENT);
if (tmp) {
*flags |= tmp;
*flags &= ~tmp;
goto next;
}
+ tmp = ll_set_opt("always_ping", s1, LL_SBI_ALWAYS_PING);
+ if (tmp) {
+ *flags |= tmp;
+ goto next;
+ }
LCONSOLE_ERROR_MSG(0x152, "Unknown option '%s', won't mount.\n",
s1);
RETURN(-EINVAL);
if (err)
GOTO(out_free, err);
lsi->lsi_flags |= LSI_BDI_INITIALIZED;
+#ifdef HAVE_BDI_CAP_MAP_COPY
lsi->lsi_bdi.capabilities = BDI_CAP_MAP_COPY;
+#else
+ lsi->lsi_bdi.capabilities = 0;
+#endif
err = ll_bdi_register(&lsi->lsi_bdi);
if (err)
GOTO(out_free, err);
sbi->ll_client_common_fill_super_succeeded = 1;
out_free:
- if (md)
- OBD_FREE(md, strlen(lprof->lp_md) + instlen + 2);
- if (dt)
- OBD_FREE(dt, strlen(lprof->lp_dt) + instlen + 2);
- if (err)
- ll_put_super(sb);
- else if (sbi->ll_flags & LL_SBI_VERBOSE)
- LCONSOLE_WARN("Mounted %s\n", profilenm);
+ if (md)
+ OBD_FREE(md, strlen(lprof->lp_md) + instlen + 2);
+ if (dt)
+ OBD_FREE(dt, strlen(lprof->lp_dt) + instlen + 2);
+ if (lprof != NULL)
+ class_put_profile(lprof);
+ if (err)
+ ll_put_super(sb);
+ else if (sbi->ll_flags & LL_SBI_VERBOSE)
+ LCONSOLE_WARN("Mounted %s\n", profilenm);
- OBD_FREE_PTR(cfg);
- RETURN(err);
+ OBD_FREE_PTR(cfg);
+ RETURN(err);
} /* ll_fill_super */
void ll_put_super(struct super_block *sb)
return inode;
}
-static void ll_dir_clear_lsm_md(struct inode *inode)
+void ll_dir_clear_lsm_md(struct inode *inode)
{
struct ll_inode_info *lli = ll_i2info(inode);
LTIME_S(inode->i_ctime) = 0;
inode->i_rdev = 0;
+#ifdef HAVE_BACKING_DEV_INFO
/* initializing backing dev info. */
inode->i_mapping->backing_dev_info =
&s2lsi(inode->i_sb)->lsi_bdi;
+#endif
inode->i_op = &ll_dir_inode_operations;
inode->i_fop = &ll_dir_operations;
lli->lli_fid = *fid;
/* set the directory layout */
if (lli->lli_lsm_md == NULL) {
+ struct cl_attr *attr;
rc = ll_init_lsm_md(inode, md);
if (rc != 0)
RETURN(rc);
- lli->lli_lsm_md = lsm;
- /* set lsm_md to NULL, so the following free lustre_md
+ /* set md->lmv to NULL, so the following free lustre_md
* will not free this lsm */
md->lmv = NULL;
+ lli->lli_lsm_md = lsm;
+
+ OBD_ALLOC_PTR(attr);
+ if (attr == NULL)
+ RETURN(-ENOMEM);
+
+ /* validate the lsm */
+ rc = md_merge_attr(ll_i2mdexp(inode), lsm, attr,
+ ll_md_blocking_ast);
+ if (rc != 0) {
+ OBD_FREE_PTR(attr);
+ RETURN(rc);
+ }
+
+ if (md->body->mbo_valid & OBD_MD_FLNLINK)
+ md->body->mbo_nlink = attr->cat_nlink;
+ if (md->body->mbo_valid & OBD_MD_FLSIZE)
+ md->body->mbo_size = attr->cat_size;
+ if (md->body->mbo_valid & OBD_MD_FLATIME)
+ md->body->mbo_atime = attr->cat_atime;
+ if (md->body->mbo_valid & OBD_MD_FLCTIME)
+ md->body->mbo_ctime = attr->cat_ctime;
+ if (md->body->mbo_valid & OBD_MD_FLMTIME)
+ md->body->mbo_mtime = attr->cat_mtime;
+
+ OBD_FREE_PTR(attr);
+
CDEBUG(D_INODE, "Set lsm %p magic %x to "DFID"\n", lsm,
lsm->lsm_md_magic, PFID(ll_inode2fid(inode)));
RETURN(0);
/* inode size will be in ll_setattr_ost, can't do it now since dirty
* cache is not cleared yet. */
op_data->op_attr.ia_valid &= ~(TIMES_SET_FLAGS | ATTR_SIZE);
+ if (S_ISREG(inode->i_mode))
+ mutex_lock(&inode->i_mutex);
rc = simple_setattr(dentry, &op_data->op_attr);
+ if (S_ISREG(inode->i_mode))
+ mutex_unlock(&inode->i_mutex);
op_data->op_attr.ia_valid = ia_valid;
rc = ll_update_inode(inode, &md);
struct inode *inode = dentry->d_inode;
struct ll_inode_info *lli = ll_i2info(inode);
struct md_op_data *op_data = NULL;
- bool file_is_released = false;
int rc = 0;
ENTRY;
}
/* We mark all of the fields "set" so MDS/OST does not re-set them */
- if (attr->ia_valid & ATTR_CTIME) {
- attr->ia_ctime = CFS_CURRENT_TIME;
+ if (!(attr->ia_valid & ATTR_CTIME_SET) &&
+ (attr->ia_valid & ATTR_CTIME)) {
+ attr->ia_ctime = CURRENT_TIME;
attr->ia_valid |= ATTR_CTIME_SET;
}
if (!(attr->ia_valid & ATTR_ATIME_SET) &&
(attr->ia_valid & ATTR_ATIME)) {
- attr->ia_atime = CFS_CURRENT_TIME;
+ attr->ia_atime = CURRENT_TIME;
attr->ia_valid |= ATTR_ATIME_SET;
}
if (!(attr->ia_valid & ATTR_MTIME_SET) &&
(attr->ia_valid & ATTR_MTIME)) {
- attr->ia_mtime = CFS_CURRENT_TIME;
+ attr->ia_mtime = CURRENT_TIME;
attr->ia_valid |= ATTR_MTIME_SET;
}
LTIME_S(attr->ia_mtime), LTIME_S(attr->ia_ctime),
cfs_time_current_sec());
- /* We always do an MDS RPC, even if we're only changing the size;
- * only the MDS knows whether truncate() should fail with -ETXTBUSY */
-
- OBD_ALLOC_PTR(op_data);
- if (op_data == NULL)
- RETURN(-ENOMEM);
-
- if (!S_ISDIR(inode->i_mode)) {
+ if (S_ISREG(inode->i_mode)) {
if (attr->ia_valid & ATTR_SIZE)
inode_dio_write_done(inode);
mutex_unlock(&inode->i_mutex);
}
- /* truncate on a released file must failed with -ENODATA,
- * so size must not be set on MDS for released file
- * but other attributes must be set
- */
- if (S_ISREG(inode->i_mode)) {
- struct cl_layout cl = {
- .cl_is_released = false,
- };
- struct lu_env *env;
- int refcheck;
- __u32 gen;
-
- rc = ll_layout_refresh(inode, &gen);
- if (rc < 0)
- GOTO(out, rc);
-
- /* XXX: the only place we need to know the layout type,
- * this will be removed by a later patch. -Jinshan */
- env = cl_env_get(&refcheck);
- if (IS_ERR(env))
- GOTO(out, rc = PTR_ERR(env));
-
- rc = cl_object_layout_get(env, lli->lli_clob, &cl);
- cl_env_put(env, &refcheck);
- if (rc < 0)
- GOTO(out, rc);
-
- file_is_released = cl.cl_is_released;
-
- if (!hsm_import && attr->ia_valid & ATTR_SIZE) {
- if (file_is_released) {
- rc = ll_layout_restore(inode, 0, attr->ia_size);
- if (rc < 0)
- GOTO(out, rc);
+ /* We always do an MDS RPC, even if we're only changing the size;
+ * only the MDS knows whether truncate() should fail with -ETXTBUSY */
- file_is_released = false;
- ll_layout_refresh(inode, &gen);
- }
+ OBD_ALLOC_PTR(op_data);
+ if (op_data == NULL)
+ GOTO(out, rc = -ENOMEM);
- /* If we are changing file size, file content is
- * modified, flag it. */
- attr->ia_valid |= MDS_OPEN_OWNEROVERRIDE;
- spin_lock(&lli->lli_lock);
- lli->lli_flags |= LLIF_DATA_MODIFIED;
- spin_unlock(&lli->lli_lock);
- op_data->op_bias |= MDS_DATA_MODIFIED;
- }
+ if (!hsm_import && attr->ia_valid & ATTR_SIZE) {
+ /* If we are changing file size, file content is
+ * modified, flag it. */
+ attr->ia_valid |= MDS_OPEN_OWNEROVERRIDE;
+ op_data->op_bias |= MDS_DATA_MODIFIED;
+ ll_file_clear_flag(lli, LLIF_DATA_MODIFIED);
}
- memcpy(&op_data->op_attr, attr, sizeof(*attr));
+ op_data->op_attr = *attr;
rc = ll_md_setattr(dentry, op_data);
if (rc)
GOTO(out, rc);
- /* RPC to MDT is sent, cancel data modification flag */
- if (rc == 0 && (op_data->op_bias & MDS_DATA_MODIFIED)) {
- spin_lock(&lli->lli_lock);
- lli->lli_flags &= ~LLIF_DATA_MODIFIED;
- spin_unlock(&lli->lli_lock);
- }
-
- if (!S_ISREG(inode->i_mode) || file_is_released)
+ if (!S_ISREG(inode->i_mode) || hsm_import)
GOTO(out, rc = 0);
if (attr->ia_valid & (ATTR_SIZE |
ATTR_ATIME | ATTR_ATIME_SET |
- ATTR_MTIME | ATTR_MTIME_SET)) {
+ ATTR_MTIME | ATTR_MTIME_SET |
+ ATTR_CTIME | ATTR_CTIME_SET)) {
/* For truncate and utimes sending attributes to OSTs, setting
* mtime/atime to the past will be performed under PW [0:EOF]
* extent lock (new_size:EOF for truncate). It may seem
* time de-synchronization between MDT inode and OST objects */
rc = cl_setattr_ost(lli->lli_clob, attr, 0);
}
+
+ /* If the file was restored, it needs to set dirty flag.
+ *
+ * We've already sent MDS_DATA_MODIFIED flag in
+ * ll_md_setattr() for truncate. However, the MDT refuses to
+ * set the HS_DIRTY flag on released files, so we have to set
+ * it again if the file has been restored. Please check how
+ * LLIF_DATA_MODIFIED is set in vvp_io_setattr_fini().
+ *
+ * Please notice that if the file is not released, the previous
+ * MDS_DATA_MODIFIED has taken effect and usually
+ * LLIF_DATA_MODIFIED is not set(see vvp_io_setattr_fini()).
+ * This way we can save an RPC for common open + trunc
+ * operation. */
+ if (ll_file_test_and_clear_flag(lli, LLIF_DATA_MODIFIED)) {
+ struct hsm_state_set hss = {
+ .hss_valid = HSS_SETMASK,
+ .hss_setmask = HS_DIRTY,
+ };
+ int rc2;
+
+ rc2 = ll_hsm_state_set(inode, &hss);
+ /* truncate and write can happen at the same time, so that
+ * the file can be set modified even though the file is not
+ * restored from released state, and ll_hsm_state_set() is
+ * not applicable for the file, and rc2 < 0 is normal in this
+ * case. */
+ if (rc2 < 0)
+ CDEBUG(D_INFO, DFID "HSM set dirty failed: rc2 = %d\n",
+ PFID(ll_inode2fid(inode)), rc2);
+ }
+
EXIT;
out:
if (op_data != NULL)
ll_finish_md_op_data(op_data);
- if (!S_ISDIR(inode->i_mode)) {
+ if (S_ISREG(inode->i_mode)) {
mutex_lock(&inode->i_mutex);
if ((attr->ia_valid & ATTR_SIZE) && !hsm_import)
inode_dio_wait(inode);
}
if (body->mbo_valid & OBD_MD_TSTATE) {
+ /* Set LLIF_FILE_RESTORING if restore ongoing and
+ * clear it when done to ensure to start again
+ * glimpsing updated attrs
+ */
if (body->mbo_t_state & MS_RESTORE)
- lli->lli_flags |= LLIF_FILE_RESTORING;
+ ll_file_set_flag(lli, LLIF_FILE_RESTORING);
+ else
+ ll_file_clear_flag(lli, LLIF_FILE_RESTORING);
}
return 0;
/* OIDEBUG(inode); */
- /* initializing backing dev info. */
- inode->i_mapping->backing_dev_info = &s2lsi(inode->i_sb)->lsi_bdi;
-
-
+#ifdef HAVE_BACKING_DEV_INFO
+ /* initializing backing dev info. */
+ inode->i_mapping->backing_dev_info = &s2lsi(inode->i_sb)->lsi_bdi;
+#endif
if (S_ISREG(inode->i_mode)) {
struct ll_sb_info *sbi = ll_i2sbi(inode);
inode->i_op = &ll_file_inode_operations;
* 2. layout was changed by another client
* 3. proc2: refresh layout and layout lock granted
* 4. proc1: to apply a stale layout */
- if (it != NULL && it->d.lustre.it_lock_mode != 0) {
+ if (it != NULL && it->it_lock_mode != 0) {
struct lustre_handle lockh;
struct ldlm_lock *lock;
- lockh.cookie = it->d.lustre.it_lock_handle;
+ lockh.cookie = it->it_lock_handle;
lock = ldlm_handle2lock(&lockh);
LASSERT(lock != NULL);
if (ldlm_has_layout(lock)) {
char *buf = NULL;
struct obd_ioctl_data *data = NULL;
__u32 type;
- __u32 __user flags; /* not user, but obd_iocontrol is abused */
int len = 0, rc;
if (!inode || !(sbi = ll_i2sbi(inode)))
else
GOTO(out_statfs, rc = -ENODEV);
- flags = (type & LL_STATFS_NODELAY) ? OBD_STATFS_NODELAY : 0;
- rc = obd_iocontrol(IOC_OBD_STATFS, exp, len, buf, &flags);
+ rc = obd_iocontrol(IOC_OBD_STATFS, exp, len, buf, NULL);
if (rc)
GOTO(out_statfs, rc);
out_statfs:
op_data->op_fsuid = from_kuid(&init_user_ns, current_fsuid());
op_data->op_fsgid = from_kgid(&init_user_ns, current_fsgid());
op_data->op_cap = cfs_curproc_cap_pack();
- op_data->op_bias = 0;
- op_data->op_cli_flags = 0;
if ((opc == LUSTRE_OPC_CREATE) && (name != NULL) &&
filename_is_volatile(name, namelen, &op_data->op_mds)) {
op_data->op_bias |= MDS_CREATE_VOLATILE;
}
op_data->op_data = data;
- /* When called by ll_setattr_raw, file is i1. */
- if (LLIF_DATA_MODIFIED & ll_i2info(i1)->lli_flags)
- op_data->op_bias |= MDS_DATA_MODIFIED;
-
return op_data;
}
if (sbi->ll_flags & LL_SBI_USER_FID2PATH)
seq_puts(seq, ",user_fid2path");
+ if (sbi->ll_flags & LL_SBI_ALWAYS_PING)
+ seq_puts(seq, ",always_ping");
+
RETURN(0);
}
git://git.whamcloud.com / fs / lustre-release.git / blobdiff